Continued drinking is common, but risky for individuals in remission from alcohol use disorder

Though many factors may intuitively increase risk for relapse to alcohol use disorder, little population level data exists on the topic. This study looked especially at history of alcohol use disorder severity, and current drinking as relapse risk factors.

WHAT PROBLEM DOES THIS STUDY ADDRESS?

Though many factors may intuitively foreshadow or predict relapse to alcohol use disorder (e.g., greater addiction severity history, current alcohol use), little population level data exists on the topic. Using a nationally representative Dutch sample, this study sought to identify the risk factors for relapse to active alcohol use disorder by surveying and clinically interviewing 506 individuals with greater than one year of recovery from alcohol use disorder, and followed them for three years.

HOW WAS THIS STUDY CONDUCTED?

Data were obtained from the first two waves of the Netherlands Mental Health Survey and Incidence Study-2 (NEMESIS-2); a longitudinal, epidemiological survey of the general Dutch population. Baseline data for the present study were collected between November 2007 and July 2009. The NEMESIS-2 survey randomly sampled Dutch households, selecting one participant per household, resulting in a total sample of 6646 adults ages 18-64. To compile the present study sample, the authors identified 506 NEMESIS-2 participants who across their lifespan endorsed at least two criteria for alcohol use disorder, who were in alcohol use disorder remission (note, to receive a formal diagnosis of alcohol use disorder one must endorse at least two symptoms within a 12-month period). This means participants had at least 12 months since experiencing these alcohol use disorder symptoms. Since alcohol use is not a diagnostic criterion for alcohol use disorder, it does not necessarily mean individuals had stopped drinking alcohol. Of the 506 individuals interviewed at baseline, 421 (83.2 percent) were available at three-year follow-up.

READ MORE ON STUDY METHODS

Alcohol use disorder was assessed using the Concordance of the Composite International Diagnostic Interview Version 3.0 (CIDI 3.0) using Diagnostic And Statistical Manual of Mental Disorder 5 alcohol use disorder criteria. To meet criteria for alcohol use disorder someone must have at least two of 11 possible symptoms occurring in a 12-month period at some point in the life-span. For the present study however, the authors simply looked at lifetime number of alcohol use disorder symptoms and the age at which symptoms first and last occurred. Aligning with Diagnostic And Statistical Manual of Mental Disorder 5, the authors classified participants as having mild (2-3 symptoms), moderate (4-5 symptoms), or severe (6+ symptoms) alcohol use disorder.

Alcohol use disorder relapse was deemed to occur if participants reported two or more alcohol use disorder symptoms between study baseline and 3-year follow-up.

Previous alcohol consumption was assessed by the question, “Think about the years in your life when you drank the most. During those years how often did you usually have at least one drink: every day, nearly every day, 3-4 days a week, 1-2 days a week, 1-3 days a month, or less than once a month?” The authors then created a categorical variable of past alcohol use that included low-risk drinking (≤14/21 drinks weekly for women/men), medium-risk drinking (15-28/22-24 drinks weekly for women/men), and high-risk drinking (≥29/43 drinks weekly for women/men). Participants also answered similar questions about past 12-month alcohol use to assess current alcohol intake. Current at-risk drinking was set at ≤8/15 drinks weekly for women/men).

WHAT DID THIS STUDY FIND?

Relapse rates

Between baseline and three-year follow-up, 46 respondents developed two or more Diagnostic And Statistical Manual of Mental Disorders 5 alcohol use disorder symptoms and were thus deemed to have relapsed to active alcohol use disorder. Their time to relapse was based on participants’ length of alcohol use disorder remission at study baseline. The cumulative relapse rate was 1.4 percent at one year, 2.9 percent at two years, 5.6 percent at five years, 9.1 percent at 10 years, and 12 percent at 20 years. Thereafter, remission appeared quite stable with a total cumulative relapse rate of 12.8 percent after 22 years. In other words, by 22 years of remission, approximately 13 percent had relapsed.

Graph showing time to alcohol use disorder (AUD) relapse among study participants. Time to relapse (horizontal axis) is measured in years from the date participants achieved AUD remission. The graph shows that risk of relapse is fairly constant up to 12 years of remission, and then drops to around 13%, and remains constant after 22 years of remission.

Alcohol use disorder characteristics as predictors of relapse

Respondents had a mean past alcohol consumption level of approximately 50 drinks weekly, and reported on average 3.4 lifetime alcohol use disorder symptoms. Upon assessment, the majority of the respondents (71.6 percent) met diagnostic criteria for mild alcohol use disorder (2-3 symptoms), 14 percent had moderate alcohol use disorder (4–5 symptoms) and 14.4 percent had severe alcohol use disorder (6+ symptoms).

Analyses showed that both medium (15–28/22–42 drinks weekly for women/men) and high (≥ 29/43 drinks weekly for women/men) levels of alcohol use while participants were in active alcohol use disorder, were associated with greater likelihood of relapse after a period of alcohol use disorder remission. Having a history of severe alcohol use disorder also was associated with greater likelihood of relapse, but lifetime moderate alcohol use disorder was not.

Regarding lifetime alcohol use disorder symptoms endorsed by participants, prevalence was highest for the alcohol use disorder symptoms ‘drinking larger quantities or longer than intended’ and ‘hazardous use’, and lowest for ‘important activities given up or reduced as a result of drinking’ and ‘alcohol use despite physical or psychological harm’.

Risk of relapse after a period of alcohol use disorder remission was higher for those endorsing the alcohol use disorder symptoms ‘unsuccessful efforts to cut back or quit’, ‘continued alcohol use despite social or interpersonal problems’, and ‘alcohol craving’.

When all variables were included in a statistical model, the diagnostic criteria ‘unsuccessful efforts to cut back or quit’ and more alcohol intake during the period of greatest drinking predicted relapse, independently of one another. This means that when considering all measures that may lead to relapse together, these two factors stood out as independently robust predictors of whether someone would relapse to active alcohol use disorder.

Current at-risk drinking

At study baseline, 31.3 percent (n= 155) of the participants, despite being in alcohol use disorder remission, nevertheless reported current at-risk drinking (i.e., ≥ 8/15 drinks weekly for women/men). Notably, the cumulative relapse rate of respondents with at-risk drinking at study baseline (25.3 percent after 20 years) was twice that of the overall cumulative relapse rate (12 percent). Current at-risk drinking was associated with greater likelihood of relapse (after accounting statistically for lifetime characteristics such as, at least medium levels of alcohol intake, lifetime history of severe alcohol use disorder, and endorsing the ‘unsuccessful efforts to cut back or quit’ diagnostic criteria). Those endorsing current at-risk drinking had five times greater likelihood of relapse in the next year compared to those not endorsing this alcohol use disorder symptom. Moreover, analyses showed that current at-risk drinking exacerbated the effect of medium to high lifetime history of alcohol intake (increasing risk nine times) and lifetime history of severe alcohol use disorder (increasing risk 25 times) on the risk of relapse. The diagnostic criteria ‘unsuccessful efforts to cut back or quit’, however, did not exacerbate this effect.

WHAT ARE THE IMPLICATIONS OF THE STUDY FINDINGS?

This study showed that only a small minority of people with lifetime alcohol use disorder who were in remission for one or more years experienced a relapse over the three-year monitoring period. The cumulative relapse rate after 20 years or remission was 12 percent. Findings suggest, however, that certain lifetime alcohol use disorder characteristics can be used as indicants of alcohol use disorder relapse risk. In particular, specific symptoms such as having ‘unsuccessful efforts to cut back or quit’ at some time in the lifespan increases risk of alcohol use disorder relapse; more than a quarter of people endorsing this diagnostic criteria relapsed within 20 years. Previous work has shown that individuals endorsing ‘unsuccessful efforts to cut back or quit’ are more likely to seek alcohol use disorder treatment, suggesting individuals meeting this diagnostic criterion are intuitively aware of this risk.

The authors also observed that past alcohol intake and number of lifetime alcohol use disorder symptoms were both associated with higher likelihood of relapse, independently of one another. Individuals with these risk factors may benefit particularly from long-term relapse prevention monitoring and/or long-term participation in mutual-aid program like Alcoholics Anonymous (AA) which has been shown to buffer against relapse risk. Other groups such as SMART Recovery may show similar benefits if people become engaged with them.

At least some alcohol use among those in alcohol use disorder remission was also a significant risk factor for relapse to active alcohol use disorder. This finding is consistent with previous research highlighting the risk of any alcohol use among those in alcohol use disorder remission. While individuals with alcohol use disorder histories are sometimes able to drink without relapsing to active alcohol use disorder (usually those with histories of mild alcohol use disorder), these findings highlight the added risk of any alcohol use among those in alcohol use disorder remission. Put simply, abstinent remission appears to be the most stable pathway for continued remission in the future.

LIMITATIONS

The authors did not diagnose alcohol use disorder in exactly the same way as the Diagnostic And Statistical Manual of Mental Disorders 5. Rather, they assessed lifetime symptoms, and did not take into consideration whether at least two of these symptoms occurred within a 12-month period.
The three year follow-up period in this study provides a limited window for relapse monitoring. It is possible this may have resulted in under-representation of actual relapse rates.
The authors note that the Concordance of the Composite International Diagnostic Interview Version 3.0 was designed and validated for the assessment of Diagnostic and Statistical Manual of Mental Disorders IV, not for Diagnostic and Statistical Manual of Mental Disorders 5. Although the criteria used in these two versions of the manual are very similar, reliability of this measure for the Diagnostic And Statistical Manual of Mental Disorders 5 has not been established. There is therefore some risk of measurement error of alcohol use disorder symptoms in this study.
In line with other population-based surveys, alcohol intake, alcohol use disorder symptoms, and alcohol use disorder remission were assessed by self-report. Reports on such behavior may be influenced by social stigma, leading people to under-report symptoms. Moreover, people may have difficulty remembering which symptoms were present, or the exact time since the last episode of alcohol use disorder. Such biases may have resulted in an under- or over-estimation of past drinking levels and lifetime alcohol use disorder symptoms.
The study focused on a small set of possible risk factors for alcohol use disorder relapse. It is possible that other risk factors (e.g., co-occurring mental disorders; history of trauma) influenced results in unknown ways.

BOTTOM LINE

For individuals & families seeking recovery: The authors’ findings suggest that individuals in alcohol use disorder remission are safest not consuming alcohol. Further, individuals who have a history of loss of control drinking and moderate to heavy alcohol use appear to have greater risk for alcohol use disorder relapse after achieving remission, and may benefit especially from engaging in long-term supportive psychotherapy, regular checkups with an addiction specialist, and/or regularly attending peer-based recovery support programs like Alcoholics Anonymous (AA).
For scientists: This work provides important population-level data on individual risk factors for alcohol use disorder relapse. Further work is also needed however, to explore more deeply individual risk factors that may confer additional risk to the factors identified in this study, such a co-occurring psychological disorders and history of trauma. Additionally, though some have begun to study relapse risk factors for ‘low-risk’ drinkers; more work is needed in this area.
For policy makers: The authors highlight important risk factors for alcohol use disorder relapse. These findings speak to the need for long-term management of alcohol use disorder, even after individuals have achieved years of sustained remission from this disorder. Ensuring access to relapse prevention services in the community will help reduce relapse rates, and in doing so reduce disease burden on healthcare systems and the broader economy.
For treatment professionals and treatment systems: This work provides important data speaking to alcohol use disorder relapse risk associated with individual factors, including a history of loss of control drinking, history of moderate to heavy alcohol use, and current at-risk alcohol use. Patients with these risk factors may especially benefit from long-term relapse prevention support including services such as recovery check-ups, supportive psychotherapy, and/or regular attendance at 12-Step programs like Alcoholics Anonymous (AA).